Thursday, 18 October 2018

Wily With Miley, And Fibonacci Frivolity

The phrase "it's better to be lucky than good" comes to mind after my prediction yesterday that:
If Brewers pitcher Wade Miley doesn't last long, the Over 7.0 looks good as well as a Dodgers win. If Miley goes deep, then it's Unders and Brewers.
While my prediction was spot on, Dodgers won, Overs was a push, how I got the result was not.

Miley didn't last long at all. He faced one batter who he walked, and was then pulled, a planned move by the manager to influence the Dodgers line-up, and Miley will now start game 6. This also happened in the 1924 World Series if anyone likes their baseball history. 
The effective starter was right-handed Brandon Woodruff, who lasted 5.1 innings before the Brewers turned to the bullpen. In NLCS series tied at 2:2, the Game 5 winners go on to win the series 70% of the time, so the chances are that Kershaw has at least one more Dodgers starting appearance in his future. 

From time to time, Pinnacle Sports have an informative article. The "What is the Fibonacci betting system?" article is not one of them.

It's also not one of their most original articles, apparently plagiarised from an October 2013 article by Kael Mansfield which appeared on the Intelligent Betting Tips website with the title "Soccer betting system Fibonacci strategy", although the author may well have published it in two places. 

The basic premise of both the original article and the plagiarised one is whether you can make a profit backing the draw in football by staking using the Fibonacci sequence, the one qualifier being that you only:
Bet on a soccer draw when the statistical probability is above 2.618
Never mind that probability is always a number between 0 and 1. 

I think the author(s) both meant that the price of the draw must be 2.62 or higher, which in itself is silly because the draw is always above this price unless the match is being fixed.

The Pinnacle article says:
Interestingly, the odds for a potential draw in all 380 ties were above the 2.618 threshold suggested as the lower limit by Archontakis and Osborne.
"Interestingly"! Show me a match where the Draw odds are anywhere close to 2.62! This is not interesting, it's mathematics. 

The article makes the claim that:
The betting principle here comes from the theory that the soccer draw is the most difficult for online sportsbooks and bookmakers to predict the outcome…thus its much easier to be exploited.
The missing apostrophe is one clue as to the level of intelligence on display here. Another is the complete nonsense of that sentence. 

Yes, for reasons previously detailed in this blog and dating back to at least 1999 and Derek McGovern's book “On Sports Betting …And How To Make It Pay“, the Draw often offers value, but there is no betting principle on display here based on using a progressive staking system, whether it has a catchy name or not. 

The article goes on:
Like other similar sports betting strategies, the principle holds that as long as you continually increase your stake, any win will overcome the previous losses.
This is not an accurate statement at all. "Any win" won't work - your win needs to be at a high enough price to cover previous losses, but the main problem with Fibonacci and progressive staking systems in general is that you soon find yourself betting uncomfortably large amounts of money. 

When you are backing Draws, with a probability of 0.25 to 0.33, over a sequence of 1,000 bets you are almost certainly going to hit a run of 20 losses.

A recent example is the Big 6 Draws which ran into a losing sequence of 14 bets.

Q: What would have been the 15th bet using Fibonacci? 

A: 610 units

How many of us in practice would be capable of placing a stake of 610 times our initial stake? Especially given that our losses to this point are 'only' 986 points. As with a bad trade, there's no need to win it all back in one bet. 

And a 21st bet would be 10,946 units.

It might work on paper, but as the saying goes, paper trading isn't worth the paper it's written on.

The original article does at least conclude that:
the Fibonacci is not a long term profitable investment strategy.
but most of us knew that already. Pinnacle do themselves a disservice by including this nonsense in their library of articles.

No comments: